A study involving thousands of staffers and players from most Major League Baseball teams revealed that 0.7 percent of those who completed test kits had antibodies for covid-19, the disease caused by the novel coronavirus, indicating that they had been infected at a previous time.

That number is greater than the rate of reported cases, approximately 0.4 percent, among the overall U.S. population. However, most of the country has not been tested, either for antibodies or in point-of-care diagnostic tests, making the MLB study an important and unprecedented “snapshot,” researchers said Sunday.

“In one sense, it suggests that only a small fraction of the population has been infected,” Jay Bhattacharya, a medical researcher at Stanford University, told reporters in a conference call. Still, he noted, covid-19 is “also further along than what you might expect by just looking at case numbers.”

The study reportedly involved 26 of the 30 MLB teams and based its results on 5,603 completed tests and surveys. Researchers said 60 people tested positive for covid-19 antibodies, and after controlling for an expected amount of false positives and negatives, that number was adjusted to 42. There were zero deaths among the test group.

Researchers noted that the test group was not perfectly “representative of the American population at large,” because it skewed toward subjects who mostly ranged in age from 20 to 64 and who generally were of elevated socioeconomic status.

The study was described as the first of its kind in terms of its national scope, and Bianca Mulaney, a Stanford medical student who is the lead author on the study, made a point of crediting major league teams for “an amazing feat” in having been able to “orchestrate the test collection in such a short period of time.”

Daniel Eichner, who runs the Sports Medicine Research and Testing Laboratory and who helped facilitate the study, emphasized that this was “a public health study that just happened to utilize MLB employees” and was not meant to provide guidance about when baseball or other sports might be able to resume in the United States.

He added that if the 0.7 percent rate of infection appeared encouragingly low, that could show the effectiveness of social distancing and the use of personal protective equipment. He said it could also mean such measures have “worked well, and we probably have to keep doing it.”

Bhattacharya said the 0.7 percent figure was surprising in part because players and team staffers would have been grouped together in spring training, in some cases in an environment featuring a higher than normal number of respiratory droplets, a few weeks before tests were completed April 14 and 15.

However, researchers pointed to the fact that the “overwhelming majority” of test subjects were not athletes but other types of team employees, from front-office executives to stadium vendors.

It takes about six to 10 days for 50 percent of people to develop antibodies, Bhattacharya said, and thus the snapshot provided by the study would have been from early April. MLB suspended spring training in mid-March, and it hopes to resume in June before starting to play games in July.

Bhattacharya said the study provided good news in the zero deaths but bad news by indicating that “the epidemic has not gotten very far.” Another positive takeaway could be gleaned from the fact that 70 percent of the subjects who tested positive described themselves as asymptomatic.

Researchers were loath to draw analysis from results from different locations, except to acknowledge that the rate of antibodies was higher in states known to have a relatively greater prevalence of the coronavirus, such as New York, as opposed to the Midwest. According to the Athletic, the Los Angeles Angels and the two New York teams had the highest rates of infection. The MLB test subjects showed lower rates, per the researchers, than in surrounding areas where similar antibody studies had been done.

The MLB study has not yet been subject to peer review, but Bhattacharya said he plans to have it undergo that process and prepare it for publication in the coming week. He added that it was unusual to hold a news conference about a study before it was peer reviewed, but he wanted to respond to what he said was a “huge amount of public interest.”

Read more: